The usage of CAN-Bus messages for engine power determination
Transcript of The usage of CAN-Bus messages for engine power determination
138
The usage of CAN-Bus messages for engine power determination
J. Čupera, T. Šmerda
Department of Engineering and Automobile Transport, Faculty of Agronomy, Mendel University in Brno, Brno, Czech Republic
Abstract
Čupera J., Šmerda T., 2010. The usage of CAN-Bus messages for engine power determination. Res. Agr. Eng., 56: 138–146.
Determination of the actual power of tractor’s engine in the operation can be done by calculation which requires the use of a range of parameters such as coefficient of rolling resistance, mechanical efficiency, the moments of inertia, etc. Their values are usually tabulated and therefore the engine power cannot be determine without simplifying. Another solution is to use the actual engine torque message from the CAN-Bus, which brings a specific value of the actual torque. The aim is to use the current torque to calculate the engine power in the deployment of tractor’s set in transport operation. The results show that at a uniform movement of the flat section the engine power reached 73 kW. When driving uphill, the value of the actual power reached from 130 to 150 kW depending on the selected gear. Using the actual parameters of torque makes possible to identify the known full speed characteristics of the current engine without the need for assembly demanding measurement techniques.
Keywords: actual engine torque; engine; tractor; engine power; CAN-Bus
Supported by the Ministry of Education, Youth and Sports of the Czech Republic, Project No. COST 356 OC191 and Internal Grant Agency TP3/2010.
Vol. 56, 2010, No. 4: 138–146 Res. Agr. Eng.
The tractor is more often used for working opera-tions during the year. The manufactures try to re-duce effort to driving or operating the tractor. The way how to affect a need for driving is the usage of components such as suspension of front axle, cab and operator’s seat, the location of controls and indi-cators. The second possibility may be improvement of performance, e.g. increase in engine power, auto-matic shifting, etc. From a technical point of view it seems to be interesting to assess the performance of the engine in the transport of specific sets.
Theoretical analysis: All procedures for the de-termination of the actual power are based on the calculation of the effects of external forces applied to vehicle. The development of electronic control
for tractors with an option to use current infor-mation about the activities of selected functional components of the CAN-Bus can be used as a base of input data of analysis (Suvinen, Saarilahti 2006). This network brings huge information about operation, for example the current torque, the torque or engine load. Whether the parameters are associated with specific torque measured on the dynamometer, it may provide engine power under various conditions in operation.
Figs 1–3 bring schematic representation of the three typical situations which may occur during transport. The simplest case would be run on the flat. Driving force brought to the driving axle must be in balance with the resistance expressed in Eq. 4.
139
Res. Agr. Eng. Vol. 56, 2010, No. 4: 138–146
Fig. 1. Force and torque balance of the tractor when driving on the flat road in an unequal movement
Fig. 2. Force and torque balance of the tractor when driving uphill in an unequal movement
Fig. 3. Force and torque balance of the tractor when driving down the hill in an unequal movement
140
Vol. 56, 2010, No. 4: 138–146 Res. Agr. Eng.
(1)
(2)
(3)
where:Fv – rolling resistance (N)Fh – driving force (N) Y – normal force (N) Gt – tractor gravity (N)Gd – gravity transferred from trailer to tractor (N)Mv – torque of rolling resistance (N m)Mh – driven torque (N m)f – coefficient of rolling resistance (–)
Driving force can be calculated from Eq. 1 as a sum of out effects.
(4)
(5)
Another form of equation includes knowledge of engine torque, ratio gear, mechanical efficiency and dynamic diameter of a wheel.
(6)
where:Mt – engine toque (N m)ic – ration gear (–) ηm – mechanical efficiency (–) rd2 – dynamic diameter of driving wheel (m)
Engine torque can be rewritten according to Eqs 5 and 6 as follows:
(7)
The calculation of engine torque brings difficulties due to coefficient of rolling resistance and mechani-cal efficiency. Both parameters are already tabulated, but especially the value of the mechanical efficiency depends on used references or experimental results of different authors. Common mechanical efficiency at spur gears is according to the implementation of at least 0.98, at 0.97 bevel wheel shaft in the case of 0.995.
(8)
where:η
čkm – efficiency of spur gearing (–)
ηkkn
– efficiency of bevel gearing (–)η
uhp
– efficiency of shaft support (–)
The overall mechanical efficiency according to Se-metko lies between 90% and 94%, Remus, Culshaw present 88%, Gega from 85% to 95%, Reiter be-tween 80% and 88%, Grečenko writes 85–90%, and Okamoto et al. 72% to 80% (Tinker 1993). It means that the accuracy of the calculation of load will be determined by the value of the mechanical efficien-cy and coefficient of rolling resistance. Therefore, a better offer is monitoring of the actual torque from the CAN-Bus. When driving uphill, the calculation of the driving forces is difficult, since it appears that resistance rises and the effects of rotary inertia and sliding mass. Force and torque balance are shown in Fig. 2.
(9)
(10)
(11)
The balance of power can be calculated by the value of the driving forces. The calculation is dif-ficult to realize in order to find the effects of rotary inertia mass connected with the driving axle.
(12)
where:Fa1 – inertia force of front axle (N)Fa2 – inertia force of turning parts (N)Fat – inertia force of tractor (N) Fan – inertia force of trailer (N)Fsn – drag force of trailer (N)Ma – inertia torque (N m)Fac – total inertia force (N)Fvc – total force of rolling resistance (N)Fsc – total force of climbing drag (N)
1 21
0; 0xi v v h vni
F F F F F
(N)
1 21
0; 0yi t di
F Y Y G G
(N)
h t d n cF G G G f G f (N)
2
t c mh
d
M iFr
(N)
2c dt
c m
G f rMi
(N m)
m n pm k kk uh (–)
1 1 21
2
0; sin( )
0 (N)
xi a v at t ai
v h vn sn an
F F F F G F
F F F F F
1 21
0; cos 0 yi t ni
F Y Y G G
(N)
1 21
2 1
0; cos
sin( )0 (N m)
oi t v v ati
t h a a d
vn an sn
M G b M M F z
G z M M M G cF h F h F h
1 2
1 2
sin (N)
h a a at an t
sn v v vn ac vc sc
F = F + F + F + F +G × ++F + F + F + F = F + F + F
1 1 21
0;
0(N m)
oi t v v hi
d vn
M G b M Y L M M
G c F h
1 2 1 1 2 2 3
3 4 4 (N)h v v vnF F F F Y f Y f Y
f Y f
141
Res. Agr. Eng. Vol. 56, 2010, No. 4: 138–146
The last case is the ride down the hill, which once again engages the inertia effect but operates in the direction of movement and reduces the value of the forces. The measurements showed that the en-gine while driving down the hill (where the trailer pushed the tractor) reduces torque at zero. In fact, the torque can be regarded as negative, leading to a change in the direction of action of the driving forces. Balance of power exists between the roll-ing resistance, driving force, inertia effects and the moving force behind a set of Fs. Force and torque balance are presented in Fig. 3.
(13)
(14)
(15)
(16)
MATeriAls ANd MeThods
Load measurement of combustion engine was realized with a support of tractor set of CASE IH Puma 195 tractor and Annaburger HTS 22.79 semi-trailer. Field measurement was done after labora-tory measuring of full speed characteristic in Vehi-cle’s laboratory of Department of Engineering and Automobile Transport (Mendel University in Brno, Czech Republic). Torque was measured on test bench equipped with Eddy current dynamometer VD 500 during partial and full supply of fuel. CAN--Bus messages were stored in the same time sam-pling; the file consisted of fuel consumption, engine
speed, fuel temperature and air temperature before and after turbocharger, torque, etc. The measured torque values together with other needed param-eters were recorded at a frequency of 18 Hz and stored in computer memory. The distance of test-ing track was 21.7 km divided into sub-sections, in particular the character of the road gradient. Total weight is given in Table 1. In the transport operation following parameters were measured or calculated: altitude, time travel, actual velocity, fuel consumption, engine speed, actual torque, fuel temperature, altitude, and location sets. Sampling frequency signals from the sensors was 5 Hz. Loca-tion tractor was determined by GPS receiver.
System of terrain measurement includes three basic parts:
data acquisition system of HBM Company,CAN-Bus monitoring through the PCMCIA CAN card of National Instruments,tractor location determined by GPS receiver.
Altitude measurement
Altitude measurement was realized using pres-sure sensors – Motorola MPX4115 – measured in the range of 15 to 115 kPa. The output signal is lin-ear over the range of 0.2 to 4.8 V. Pressure accuracy is 1.5%. Calibration of the sensor is realized on the triangulation point with defined altitude. Uncertainty of the measurement is better than 0.5 m. The sensor was powered via USB voltage stabilizer and the value of supply voltage led to the input of card for meas-urement correction. Signal was processed by the USB DAQ card from National Instruments products.
The CAN-Bus monitoring
Software allows monitoring of CAN-Bus of a trac-tor. CAN-bus of tested tractor was fully compat-ible according to standard SAE J1939 in extended ArbID (29-bit). Communication speed was 250 kbps and for the analysis of the report relevant channels were selected.
resulTs
Measured values were processed and evaluated by spreadsheet editor Microsoft Excel. Using re-
••
•
1 1 2 21
=1
0;
0 (N)
xi v a at v a vni
an h st sn
F = F + F + F + F + F +F +
+ F + F – F – F =
1 21
0; cos 0 yi t ni
F Y Y G G
(N)
1 1 11
2 2
0; – sin
0 (N m)
oi a v ti
at v a n an vn
h sn st
M M – M – Y × L+G × ×b –
– F × z – M – M – G ×c – F × h – F ×× h – M + F × h+ F × z
1 2 1 2
(N)v v vn h a a
at an st sn
F + F + F + F + F + F ++F + F = F + F
Table 1. Weight of set (in kg)
CASE IH Puma 195 + trailer Annaburger HTS 22.79 (load)
Front axle of tractor 2,380
Total weight of tractor 12,260
Tractor + trailer 31,760
Trailer 19,460
142
Vol. 56, 2010, No. 4: 138–146 Res. Agr. Eng.
gression analysis they were processed according to Mtakt = f (Mt) and Mt = f (neng) to establish the full features, Figs 4, 5. Determinacy index values ranged from 0.97 to 0.99. Measured values of transport are graphically summarized in graphs; Figs 7–9.
Table 2. Statistics of selected indicators during the plane ride
Index Velocity v (km/h) Altitude h (m) Engine speed n (1/min) Actual engine torque (%)
Averaged 41.40 210 1,745.68 49.30
Standard deviation 0.11 0.72 4.79 1.42
Minimum 41.11 209.70 1,734.00 46.00
Maximum 41.48 210.30 1,757.00 52.00
Number of values 133.00 133.00 133.00 133.00
Variation coefficient (%) 0.27 0.12 0.27 2.89
Table 3. Values used to calculation
ic (for 19th gear) (–) 14.798
rd2 (m) 0.925
ηm (–) 0.85–0.9
300
400
500
600
700
800
900
ine
torq
ue M
t(N
m),
mea
sure
d vi
a PT
O
measured values
calculated values after regression analysis
y = -6.2291n + 11,446R² = 0.9991
0
100
200
300
400
500
600
700
800
900
1,000 1,200 1,400 1,600 1,800 2,000
Engi
ne to
rque
Mt(
N m
), m
easu
red
via
PTO
Engine speed n (1/min)
measured values
calculated values after regression analysis
Fig. 4. Dependence of actual engine torque versus engine torque measured on dynamometer
Fig. 5. Dependence of actual engine torque versus engine torque measured on dynamometer, regulation part from engine performance, the CASE IH 195 Puma tractor; power take off (PTO)
y = 0.0953Mt + 11.657R² 0 999520
30405060708090
100
ctua
l eng
ine
torq
ue
(%)
mea
sure
d fr
om C
AN
-Bus
measured values
calculated values after regression analysis
y = 0.0953Mt + 11.657R² = 0.9995
0102030405060708090
100
0 200 400 600 800 1,000
Act
ual e
ngin
e to
rque
(%
)m
easu
red
from
CA
N-B
us
Engine torque Mt (Nm), measured via PTO
measured values
calculated values after regression analysis
143
Res. Agr. Eng. Vol. 56, 2010, No. 4: 138–146
90 %85 %
75 %70 %65 %60 %55 %50 %45 %40 %
Pe (kW)1571501401301201101009080
80 %
70300
400
500
600
700
800
900
ngin
e to
rqu
Mt(
N m
)
90 %85 %
75 %70 %65 %60 %55 %50 %45 %40 %
Pe (kW)1571501401301201101009080
80 %
70
0
100
200
300
400
500
600
700
800
900
1,200 1,300 1,400 1,500 1,600 1,700 1,800 1,900 2,000 2,100 2,200 2,300 2,400
Engi
ne to
rqu
Mt(
N m
)
Engine speed n (1/min)
150
170
190
210
230
250
1 000
1,300
1,600
1,900
2,200
2,500
Altitude (m
)
Engi
ne sp
eed
n(1
/min
)
Engine speed Altitude
150
170
190
210
230
250
1,000
1,300
1,600
1,900
2,200
2,500
425 430 435 440 445 450
Altitude (m
)
Engi
ne sp
eed
n(1
/min
)
Time t (s)
Engine speed Altitude
20
40
60
80
100
1,300
1,600
1,900
2,200
2,500
Actual engine torque (%
)Engi
ne sp
eed
n(1
/min
)
Engine speed Actual engine torque0
20
40
60
80
100
1,000
1,300
1,600
1,900
2,200
2,500
425 430 435 440 445 450
Actual engine torque (%
)Engi
ne sp
eed
n(1
/min
)
Time t (s)
Engine speed Actual engine torque
Fig. 6. Full speed map of the CASE IH 195 Puma tractor with constant curve of actual engine torque and power
Fig. 7. Record of measured parameters in driving on the flat road; altitude (a), actual engine torque (b)
(a)
(b)
P
144
Vol. 56, 2010, No. 4: 138–146 Res. Agr. Eng.
disCussioN
Rated speed characteristic was measured in or-der to determine the load of the combustion engine during its operation. Full speed map of engine was built of ten partial load curves (Fig. 6). The curves were plotted on constant actual torque in %, which is given by approximation using the specific values of engine torque and engine speed (Figs 4 and 5). Parameters of CAN-Bus were monitored to bring the full features and to determine the value of the actual torque and engine power. Testing track was divided into section with different profile where the load of engine varied from zero to full torque. Nine-teenth gear was shifted on a flat road. Evaluation in selected section was targeted that the movement of tractor approached balance straightforward move-ment. This is also confirmed by statistical evalua-tion of selected parameters measured during the
trip, Table 2 and Fig. 7. The location of actual en-gine torque in full speed map can be described on the basis of two major variables – engine speed and engine load (Fig. 6). Value of engine power on flat road has reached 73 kW, which corresponds to 46% of maximum power. The Eq. 7 can be calculated as the value of the coefficient of rolling resistance for the set. Parameters for calculation are listed in Table 3. The results show that the coefficient of rolling resistance for the set ranged from 0.0175 to 0.0185 for the mechanical efficiency of 0.85 and 0.9. Resulting values correspond to the range of report-ed references Marcín, Zítek (1985), Semetko et al. (1986), Grečenko (1994), Macmillan (2002). The same procedure can be used for determination of the engine power when driving uphill. The full speed map consists of three gears (16, 17, and 18), which were shifted in the selected parts of the track, Fig. 8. Four characteristic points were selected for
200
220
240
260
280
300
1,000
1,300
1,600
1,900
2,200
2,500
100 110 120 130 140 150 160 170 180 190 200
Altitude (m
)
Engi
ne s
peed
n(1
/min
)
Time t (s)
Engine speed Altitude
0
20
40
60
80
100
1,000
1,300
1,600
1,900
2,200
2,500
100 110 120 130 140 150 160 170 180 190 200
Actual engine torque (%
)Engi
ne sp
eed
n(1
/min
)
Time t (s)
Engine speed Actual engine torque
Fig. 8. Record of measured parameters in driving uphill; altitude (a), actual engine torque (b)
(a)
(b)
145
Res. Agr. Eng. Vol. 56, 2010, No. 4: 138–146
200
220
240
260
280
300
0
500
1,000
1,500
2,000
2,500
80 90 100 110 120 130 140 150
Altitude (m
)
Engi
ne sp
eed
n(1
/min
)
Time t (s)
Engine speed Altitude
0
20
40
60
80
100
0
500
1,000
1,500
2,000
2,500
80 90 100 110 120 130 140 150
Actual engine torque (%
)
Engi
ne sp
eed
n(1
/min
)
Time t (s)
Engine speed Actual engine torque
Fig. 9. Record of measured parameters in driving downhill; altitude (a), actual engine torque (b)
Table 4. Selected parameters brought up into total engine performance calculation (*average values)
Measured value Computed value from total engine performance
engine speed (1/min) actual engine torque (%) engine power (kW) engine torque (N m)
Riding on the flat 1,745* 49 73.05 400
Ride uphill 16th ration gear
1,905 80 141.76 711
1,800 80 134.89 716
1,700 81 129.17 726
1,638 80 124.12 724
Ride uphill 17th ration gear
1,900 82 143.18 720
1,800 84 142.80 758
1,700 85 137.36 772
1,635 83 129.71 758
Ride uphill 18th ration gear
1,901 84 150.02 754
1,808 84 143.25 757
1,703 85 137.60 772
1,633 83 129.55 758
Ride downhill 1,905* 0 0 0
(a)
(b)
146
Vol. 56, 2010, No. 4: 138–146 Res. Agr. Eng.
each gear, Figs 4, 5 and these points were plotted into the full speed map Fig. 6. The results show that the tractor operated with an increase in the engine power between 130 to 150 kW at 17th and 18th gear. Maximum engine power was consumed at 83% to 96%. Downshifting to 16th gear brought the reduc-tion in boost due to the higher gear ratio and over-load in transmission. Driving down the hill, actual torque message cannot be used to calculate power, as the set is pushed and the engine sets the zero fuel supply and the actual torque generated by engine is zero, Fig. 9.
CoNClusioN
The actual load can provide the value of actual engine power whether the full speed map had been measured. This procedure brings advantages in needs for installed instrumentation what is posi-tive, especially in field measurement.
r e f e r e n c e s
Grečenko A., 1994. Vlastnosti terénních vozidel (Properties of Off-Road Vehicles). 1st Ed. Prague, Czech University of Agriculture: 118.
Macmillan H.R., 2002. The Mechanics of Tractor – Implement Performance. Melbourne, University of Melbourne: 165.
Marcín J., Zítek P., 1985. Pneumatiky (Tyres). 1st Ed. Prague, SNTL – Nakladatelství technické literatury: 491.
Semetko J., Janoško I., Perniš P., 2004. Určovanie výkonu vozidla s pohonom viac náprav (Determination of Power of Multidrive Vehicles). Acta technologica agriculturae, 7: 20–23.
Semetko J., Drabant Š., Matějka J., Pick E., Šmicr V., Žikla A., 1986. Mobilní energetické prostředky (Mobile Energetic Vehicle). 1st Ed. Bratislava, Příroda: 453.
Suvinen A., Saarilahti M. 2006. Measuring the mobility parameters of forwarders using GPS and CAN Bus tech-niques. Journal of Terramechanics, 43: 237–252.
Tinker B.D., 1993. Integration of tractor engine, transmis-sion and implement depth controls. Journal of Agriculture Engineering Research, 54: 1–27.
Received for publication December 4, 2009 Accepted after corrections February 21, 2010
Corresponding author:
Ing. Jiří Čupera, Ph. D., Mendel University in Brno, Faculty of Agronomy, Department of Engineering and Automobile Transport, Zemědělská 1, 613 00 Brno, Czech Republic phone: + 420 545 132 942, e-mail: xcupera@node. mendelu.cz